Detection Of Deregulated Module In Glioblastoma Based On Integrated Network

Complex disease(e.g. cancer) is the most important factor restricting human lifespan. Moreover, it affects people’s normal life and work. Modern biomedical research has shown that both the human health and disease states have direct or indirect relationship with genes. How to treat and diagnose the complex diseases has become a difficult and hot topic in biomedical research. For a long time, since both the data analysis techniques and biological experimental techniques are not very mature, the research of complex diseases has failed to achieve effective progresses. Fortunately, the rapid development of high-throughput biotechnology provides a reliable source of data for the research of complex diseases. At the same time, the biological interaction networks generated by high-throughput techniques are powerful and useful tools for these studies, such as protein-protein interaction networks.Most of the existing studies for complex disease focus on the prediction of disease genes. However, it neglects the regulation relationship between different biomolecules. In complex diseases, various combinations of genomic perturbations or mutations often lead to the same disease and these perturbations are expected to converge to the common pathway. Therefore, focusing on the dysregulated modules rather than individual genes, we detect dysregulated modules of glioblastoma multiforme(GBM). Considering the incomplete information of single biological network, we integrate a protein-protein interaction network, a transcription-protein interaction network and a protein phosphorylation network to construct an integrated biological network. First, we identify a representative set of genes that are differentially expressed in cancer cases compared to non-tumor control cases by SAM. Then, a possible causal loci of each target gene is identified by an e QTL analysis. Finally, we detect a set of putative causal genes and a dysregulated module by utilizing a random walk method.Finally, we have successfully detected a dysregulated module consisting of 172 genes and 206 links among them. Remarkably, the gene CDKN2 A, EGFR, TP53, AKT1, RB1, PTEN and PRKCA in the dysregulated module are also reported to be closely associated with GBM in OMIM, KEGG and Ace View databases. Although the gene RHOBOTN2, CEBPA and GBAS are not recorded in these databases, we have proved that they are all closely related to GBM by the comprehensive literatures validation. Moreover, the dysregulated module contains a majority of disease genes which are identified by analyzing a single biological network. The results are validated by the statistical analysis and biological significance, indicating that the dysregulated module is closely related to GBM. Therefore, we provide a novel and effective method for detecting dysregulated modules.